Hallo, Federico hat gesagt: // Federico wrote:

I tried running pdx7 (taken from footils.org) but I get some errors:

mtx_*~ 2 6 ... couldn't create ...

mtx_*~ is the replacement object for matrix~ in the current iemmatrix as it is in CVS, so I guess, you are not running the latest iemmatrix? It also might be that you need to build and load iemmatrix as a library, because "mtx_*~ is a strange name. Maybe IOhannes knows more?

Do you have "matrix~"? You could replace [mtx_*~] with [matrix~], however then you also need to adapt the arguments and connections according to this comment from mtx_mul~.c:

/* arguments parsing:

• this might depend on whether we are creating an object
• [mtx_*~], [matrix~] or [matrix_mul_line~]
• [mtx_*~ [#out [#in [time]]]]: in1=matrix; in2-in(#in+1)=signals; in(#in+2)=time
• [matrix~ [#in [#out [time]]]]: in1-in(#in)=signals; in(#in+1)=matrix; in(#in+2)=time
• [matrix_mul_line~ [#in [#out [time]]]]: in1=matrix; in1=time; in1-in(#in):=signals
• furthermore:
• [mtx_*~] and [matrix_mul_line~] : O^=A*I^
• [matrix~] : O^'=I^'*B
• with "matrix=(A or B)" and "A=B'"

*/

Unfortunatly [mtx_*~] doesn't have a help patch.

• OSC-route: float arguments are not OK

OSCroute $1 ... couldn't create ...

If you load the help file, this shouldn't happen. rrad.pdx7~.pd is intended to become part of the RRADical collection, so you need to create it with a first argument starting with a slash to use the OSC and state-saving functionality of RRADical. If you don't supply a /argument, then you will get this error message and you will not be able to use OSC or state saving. Everything else should continue to work.

pdx7.save.pd 3 1 2 0 (canvas->outlet) connection failed

This is the [OSCroute $1] from above that wasn't created, because $1 is empty is 0 and thus a float argument that is "not OK", as far as OSCroute is concerned.

Ciao

Frank Barknecht _ ______footils.org_ __goto10.org__

Hallo, Federico hat gesagt: // Federico wrote:

however I fixed the above errors, and still get these:

pdx7.save.pd 3 1 2 0 (canvas->outlet) connection failed pdx7.save.pd 3 1 2 0 (canvas->outlet) connection failed expr, expr~, fexpr~ version 0.4 under GNU General Public License pdx7.save.pd 3 1 2 0 (canvas->outlet) connection failed pdx7.save.pd 3 1 2 0 (canvas->outlet) connection failed pdx7.save.pd 3 1 2 0 (canvas->outlet) connection failed pdx7.save.pd 3 1 2 0 (canvas->outlet) connection failed pdx7.save.pd 3 1 2 0 (canvas->outlet) connection failed pdx7.save.pd 3 1 2 0 (canvas->outlet) connection failed

hard to say, what these are: Do you also get some "couldn't create" or similar messages? Or can you open rrad.pdx7~ with "Rightclick->Open" and look for objects painted with dashed lines? One thing that could still be missing are the [list]-abs abstractions I use. They are in CVS at /abstractions/footils/list-abs/, some docs are here: http://footils.org/cms/show/46

You don't happen do already get sound, don't you?

Ciao

Frank Barknecht _ ______footils.org_ __goto10.org__

Hallo, Federico hat gesagt: // Federico wrote:

...

...

however I fixed the above errors, and still get these:

pdx7.save.pd 3 1 2 0 (canvas->outlet) connection failed pdx7.save.pd 3 1 2 0 (canvas->outlet) connection failed

hard to say, what these are: Do you also get some "couldn't create" or similar messages?

no, I don't.

Ah, now I know what these are: pdx7.save.pd is a simple abstraction wrapping the [comun] objects found in RRADical's Memento. You can find them in abstractions/rradical/memento in the CVS, if you want. They are only needed for state saving and OSC support. So would will get sounds, but you cannot easily save the current settings and you cannot use the OSC remote control yet.

(I used a wrapper around commun, because I'm also working on a version of the pdx7 which doesn't require memento, but instead uses a different saving system called "sssad". For that, only replacing pdx7.save.pd with a differnent patch will be necessary.)

Ciao

Frank Barknecht _ ______footils.org_ __goto10.org__

it was me who started collecting the synths and stuff.

..and then i got a gig in russia, and then i got a multi/o, and then i installed linux.

ok.....i will put together my collection asap

"Ideally, they would all have the same interface wherever makes sense"

this is the bit that made me lose my stride..i will have a think about the best way to do it.

On Mar 29, 2006, at 9:17 AM, carmen wrote:

...

"Ideally, they would all have the same interface wherever makes sense"

this is the bit that made me lose my stride..i will have a think
about the best way to do it.

if you mean control interface (as in OSC Schema) i agree..

if youre talking about where to place the 3 thousand sliders and
bangs, no thanks..

I meant the synth object rather than the GUI. Ideally, they would be
separate, and all of the synth objects would have a standardized
interface, so if you make a GUI for one synth object, it would be
easy to point that GUI at another synth object.

.hc

        "The arc of history bends towards justice."
                                        - Dr. Martin Luther King,  

Jr.

...

accepting data in the range of 0-1 (not 0-127! undo the MIDI
revolution!).

Except for note and velocity values, because especially for synths it makes a lot of sense to follow the midi revolution there, if just to be able to play it with a midi keyboard without having to rescale and

the 127 is probably going to be rescaled back to 0-1 or something anyway, id rather do this on input, than much more often in some synthesis code..

for frequencies 0-1 doesn't make any sense anyway.

sure it does..multiply by samplerate and divide by two..!

On Mar 29, 2006, at 10:42 AM, Frank Barknecht wrote:

Hallo, carmen hat gesagt: // carmen wrote:

...

the 127 is probably going to be rescaled back to 0-1 or something anyway, id rather do this on input, than much more often in some synthesis code..

The 127 for frequencies is scaled by [mtof] which doesn't scale to 0-1. The velocity scaling is indeed scaled to 0-1 most of the time, but personally I like being able to just connect [notein] to [oldschoolsynth] or [fluid~] directly.

0-1 does not make sense for freq, but for velocity it is a given.
That already is the standard in Pd for "velocity" i.e. amplitude.
Its also the standard for all the mapping objects, which will
complement the synth objects really nicely. Its also the standard
for Gem.

This really seems like a lot of benefits to me, especially if you
consider that you'd only have to add one object to make it work with
your simple example above:

[notein] | | | [/ 127] |
[oldschoolsynth] | [dac~]

.hc

"[W]e have invented the technology to eliminate scarcity, but we are
deliberately throwing it away to benefit those who profit from
scarcity." -John Gilmore

On Mar 29, 2006, at 1:29 PM, Martin Peach wrote:

carmen wrote:

...

...

...

accepting data in the range of 0-1 (not 0-127! undo the MIDI
revolution!).

Except for note and velocity values, because especially for
synths it makes a lot of sense to follow the midi revolution there, if just to be able to play it with a midi keyboard without having to rescale
and

the 127 is probably going to be rescaled back to 0-1 or something
anyway, id rather do this on input, than much more often in some
synthesis code..

...

for frequencies 0-1 doesn't make any sense anyway.

sure it does..multiply by samplerate and divide by two..!

Then you get different frequencies for different sample rates,
which is not usually what you want.

Yeah, that would defeat the purpose. With frequency, 0-1 does not
make sense. But a standard does. Like all synth objects using
frequency in hertz, rather than MIDI note #s, or whatever else
someone might think of.

.hc

If you are not part of the solution, you are part of the problem. -
Eldridge Cleaver

MIDI note numbers are kind of inherently logarithmic. If you want a just scale, even the decimal note numbers are kind of arbitrary. There are also scales out there that *don't* have octave symmetry. Wendy Carlos, among others, has experimented with this. There is also Georgian "quintave" singing, which uses the perfect 5th instead of the octave. And gamelan tunings are notorious for wandering octaves.

If one is using MIDI notes to trigger samples, it is possible to create soundfonts using pre-tuned samples, so you don't need decimals. But in this case it's necessary to know ahead of time what sorts of modulations you will use, which isn't always possible in extended just intonation. Not everyone uses scales. There is theoretically no limit to the number of possible notes one might use.

Someone who's interested in alternate tunings is more likely to be interested in 19-, 31-, or 53-tone equal temperament than 24. If 12 midi notes is still an octave, this will make some pretty inscrutable note numbers. If each whole note number is assigned to a different pitch of the scale, there's barely more than 2 octaves in 128 notes. Also, if you're going to try to realize something like this using pitch bends, there can only be one pitch bend value per channel, so each note has to be on a different channel than the last one. This has turned out to be a curse for me, since most soft synths don't support 16 simultaneous channels with patch changes on each note... each module is rather assigned to a specific channel.

For my purposes, frequency in Hz is the only way to go. MIDI is useful sometimes, but its inherent limits are now second nature to millions of musicians, who will never outgrow those limitations.

-Chuckk

On 3/30/06, cyborgk@nocturnalnoize.com cyborgk@nocturnalnoize.com wrote:

Hans, I have to respectfully disagree with you there. Midi notes are pretty logical to work with for most situations. Note, that this is the case only when you can have fractional note numbers, ie. 60.5 is a quarter tone up from middle C.

So, even if I wanted a non-standard scale, if it has octave symmetry, it would be easy to use MIDI note numbers. Since a wrapper exists for frequency to midi note, this shouldn't be particularly limiting. Granted, mtof allows conversion the other way, so you could make the opposite argument. But for most math manipulations, it's much easier to deal with the math of pitches divided in semitones, than to do the math needed for frequencies. If you do want to use some kind of other scale, it should be something linear, aka, 0 to 1 equals one octave wouldn't be too bad to work with.

The main limitation of MIDI, besides speed, is that CC#'s have a resolution of 128. I think that all parameters should have higher resolution, but a synth designed to play "notes" of some sort would be better off sticking to the MIDI standard.

~David

...

Yeah, that would defeat the purpose. With frequency, 0-1 does not make sense. But a standard does. Like all synth objects using frequency in hertz, rather than MIDI note #s, or whatever else someone might think of.

.hc

---

PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list

-- "It is not when truth is dirty, but when it is shallow, that the lover of knowledge is reluctant to step into its waters." -Friedrich Nietzsche, "Thus Spoke Zarathustra"

On 3/31/06, Mathieu Bouchard matju@artengine.ca wrote:

On Thu, 30 Mar 2006, Chuckk Hubbard wrote:

...

There are also scales out there that *don't* have octave symmetry.

It's (very) rare.

I suppose, but Pd is a big draw for people who want to do rare things. This 12-tET piece doesn't have octave symmetry: http://engineerc.com/~chuckk/Waking.mp3 If I'd had the resources I'd have loved to make it something other than 12-tET. Gamelan tunings almost *never* have octave symmetry. Of course, they have inharmonic overtones as well, but that's nothing Pd couldn't do.

...

Wendy Carlos, among others, has experimented with this. There is also Georgian "quintave" singing, which uses the perfect 5th instead of the octave. And gamelan tunings are notorious for wandering octaves.

Then you get nicer numbers if you multiply the MIDI note by N*log(3)/12*log(2).

You can get pretty nice numbers by not using MIDI note numbers, too.

...

Someone who's interested in alternate tunings is more likely to be interested in 19-, 31-, or 53-tone equal temperament than 24. If 12 midi notes is still an octave, this will make some pretty inscrutable note numbers.

Multiply the MIDI note number by 19/12, 31/12 or 53/12.

I guess it's not that big of a drawback, but it will be harder to tell by looking what note of the scale you have. For MIDI input, it would make more sense to multiply the incoming number by 12/19.

...

If each whole note number is assigned to a different pitch of the scale, there's barely more than 2 octaves in 128 notes.

You don't have to limit yourself to 0-127 in Pd. You also are not limited to integer note numbers.

Right, but I'm discussing the possibility of using an external sampler. I've done this in the past, simply created pre-tuned samples for those note numbers. It's one possibility, if using a sampler. The only other option with an external sampler is pitch bend, which limits you to 16 notes (seldom is there a 16-note chord, but 16 instruments or pedal points are common requirements), and requires program changes and pitch bends for each note. In my experience, external synths, hardware or software, are completely useless for any sort of polyphonic alternate tuning use, since they don't use Hz.

...

For my purposes, frequency in Hz is the only way to go. MIDI is useful sometimes,

Understand the difference between Pd's MIDI note numbers and ordinary MIDI note numbers.

I don't see why MIDI note numbers are necessary, when you can simply calculate frequency directly. They always presume that Note + 12 = Frequency * 2. For musicians who don't work that way, I see no need for it.

hi,

Am Samstag, 1. April 2006 00:31 schrieb Martin Peach:

Chuckk Hubbard wrote:

...

In my experience, external synths, hardware or software, are completely useless for any sort of polyphonic alternate tuning use, since they don't use Hz.

Older hardware works by Volts per Octave (e.g. Arp, Moog) or Volts per Hertz (e.g. Korg).

ouch .... i would recommend to use such a synth for lfo applications only ... imagine you want a 2 khz sound ... meaning 2 kilovolts ... ;-)

furthermore, on most analogue synths, you have a 0..+10 volts range for the control signals .... and some of them allow to adjust the volts/frequency ratio, giving lot of of fun, i tell you ;-)

greets,

chris

If you use a 16-bit DAC you can get a lot of frequencies. The arp2600 had a keyboard adjustment for the intervals between notes. Not polyphonic though unless you have a few of them :). Martin

---

PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list

My school has one of the first Moogs, which I managed to get my hands on, and a smaller Moog and an Arp 2600. Great fun (an amazing opportunity really), and unlimited microtonality, but not always precise. I don't know much about maintenance, it could just be that they're old, but it doesn't take long for the oscillators to start to wander. Also, not only restricted to octave equivalence, but to 12 notes. It's easy to lose a few hours to their seductive ways, though.

On 4/3/06, Martin Peach martinrp@vax2.concordia.ca wrote:

Christian Klippel wrote:

...

hi,

Am Samstag, 1. April 2006 00:31 schrieb Martin Peach:

...

Chuckk Hubbard wrote:

...

In my experience, external synths, hardware or software, are completely useless for any sort of polyphonic alternate tuning use, since they don't use Hz.

Older hardware works by Volts per Octave (e.g. Arp, Moog) or Volts per Hertz (e.g. Korg).

ouch .... i would recommend to use such a synth for lfo applications only ... imagine you want a 2 khz sound ... meaning 2 kilovolts ... ;-)

Yeah right...upside down as usual (-; I was trying to say that if you put in 1 Volt you could get (depending on your initial offset) 440Hz and then at 2 volts you get either 880Hz (an octave) on one type of analog synth or 540 Hz (440 + 100 at 100Hz/Volt) on another. That is to say, frequency in an analog synth is determined by offset+(Octave/Volt) or offset + konstant*(Hertz/Volt). Anyway, there is no absolute frequency in an analog synth the way there is in a digital one, and the ratio is usually adjustable too so alternate tunings are much easier to implement.

Martin

---

PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list

-- "It is not when truth is dirty, but when it is shallow, that the lover of knowledge is reluctant to step into its waters." -Friedrich Nietzsche, "Thus Spoke Zarathustra"

Mathieu Bouchard a écrit :

On Thu, 30 Mar 2006, Chuckk Hubbard wrote:

...

If each whole note number is assigned to a different pitch of the scale, there's barely more than 2 octaves in 128 notes.

You don't have to limit yourself to 0-127 in Pd. You also are not limited to integer note numbers.

Hey, (I just got a midi keyboard controller) why not multiplying two 0-127 ctlin to obtain more precision !! Any success story regarding that ?

...

For my purposes, frequency in Hz is the only way to go. MIDI is useful sometimes,

I find PD's midi notes very flexible. You an use them as withs scales with commas too. Look my jam abstraction on my website, maybe. (in muzaq)

Alexandre Quessy http://alexandre.quessy.net

On Sat, 1 Apr 2006, Thomas Ouellet Fredericks wrote:

I think a COARSE and FINE knob is better as on anlog synths. Example: (value of first knob * 127) + value of second knob

The shift key does a better job than a separate FINE knob.

_ _ __ ___ _____ ________ _____________ _____________________ ... | Mathieu Bouchard - tél:+1.514.383.3801 - http://artengine.ca/matju | Freelance Digital Arts Engineer, Montréal QC Canada

How do you mean? It doesn't seem necessary to do that within Pd, since you can create whatever frequency you want without need of MIDI controllers. For controlling other things with Pd, I wouldn't think you would have to do it manually with knobs, Pd can do the calculations for you.

-Chuckk

On 4/1/06, Thomas Ouellet Fredericks iamonthebeach@gmail.com wrote:

I think a COARSE and FINE knob is better as on anlog synths. Example: (value of first knob * 127) + value of second knob

Tom

On 4/1/06, Alexandre Quessy listes@sourcelibre.com wrote:

...

Mathieu Bouchard a écrit :

...

On Thu, 30 Mar 2006, Chuckk Hubbard wrote:

...

If each whole note number is assigned to a different pitch of the scale, there's barely more than 2 octaves in 128 notes.

You don't have to limit yourself to 0-127 in Pd. You also are not limited to integer note numbers.

Hey, (I just got a midi keyboard controller) why not multiplying two 0-127 ctlin to obtain more precision !! Any success story regarding that ?

...

...

For my purposes, frequency in Hz is the only way to go. MIDI is useful sometimes,

I find PD's midi notes very flexible. You an use them as withs scales with commas too. Look my jam abstraction on my website, maybe. (in muzaq)

Alexandre Quessy http://alexandre.quessy.net

---

PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list

---

PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list

-- "It is not when truth is dirty, but when it is shallow, that the lover of knowledge is reluctant to step into its waters." -Friedrich Nietzsche, "Thus Spoke Zarathustra"

Alexandre Quessy wrote:

Hey, (I just got a midi keyboard controller) why not multiplying two 0-127 ctlin to obtain more precision !! Any success story regarding that ?

The MIDI spec (see http://www.borg.com/~jglatt/tech/midispec.htm under Defined Controllers) uses the first 32 controllers as coarse values and the second 32 as the corresponding fine values for a total of 14 bits each.

Martin

Hello,

...

Then you get different frequencies for different sample rates,
which is not usually what you want.

Yeah, that would defeat the purpose. With frequency, 0-1 does not
make sense. But a standard does. Like all synth objects using
frequency in hertz, rather than MIDI note #s, or whatever else
someone might think of.

.hc

Is there an object for computing twelfth root, or anything root of two? For determining frequency difference between each interval in any tempered scale, I guess it could be useful.

Patco.

___________________________________________________________________________ Nouveau : téléphonez moins cher avec Yahoo! Messenger ! Découvez les tarifs exceptionnels pour appeler la France et l'international. Téléchargez sur http://fr.messenger.yahoo.com

this site also has soem great "standalone" tracks, each with a bunch of synths that would be easy to rip:

http://www.obiwannabe.co.uk/

(scroll down near the bottom for the patches)

I agree, this is awesome stuff. I tend to put off inventing good sounds because it seems so simple compared to the other stuff I've been doing. There are some great sounds here that I'll have to examine to learn from. Thanks for sharing!

On 4/2/06, hard off hard.off@gmail.com wrote:

this site also has soem great "standalone" tracks, each with a bunch of synths that would be easy to rip:

http://www.obiwannabe.co.uk/

(scroll down near the bottom for the patches)

---

PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list

-- "It is not when truth is dirty, but when it is shallow, that the lover of knowledge is reluctant to step into its waters." -Friedrich Nietzsche, "Thus Spoke Zarathustra"